We propose to obtain additional evidence relating levels of the O6- alkylguanine DNA alkyltransferase (AGT) to susceptibility to treatment- related acute non-lymphocytic leukemia (t-ANLL). We suppose that individuals vary in their AGT levels, that treatment of a primary malignancy with regimens including procarbazine depletes the endogenous AGT activity, and that the drug is therefore more likely to act as a carcinogen in individuals with low AGT activity, We found that individuals with ANLL de novo have the same peripheral blood lymphocyte (PBL)-AGT activities as a healthy control group but significantly higher activity than a group of t-ANLL patients. Patients on procarbazine therapy have detectable AGT activity, but this activity is lower than that of the control group. We wish to answer the following questions: 10 Is the level of AGT activity characteristic of the class of PBL, or is there great heterogeneity? 2) Does the AGT activity of control and patient groups vary in a manner that is different from the activity of other repair proteins? 3) Can we demonstrate a significant depression in AGT activity in the PLBs at the time of transformation and in Epstein-Barr virus (EBV)_ transformed lymphoblastoid lines, but both high and low activity lines have been obtained from the same individual transformed at different times. We suppose that the gene (s) controlling AGT activity may be turned on or off. We propose to determine if this is so by studying the biochemical properties of AGT prepared from high and low activity lines. We want to attempt to modulate AGT activity buy treatment with azacytidine or phorbol ester to see whether it is possible to induce activity in low-activity lymphoblastoid lines and to gain some insight as to the mode of regulation of this protein.